Open AccessCONTRIBUTIONS OF SELECTED TRANSCRIPT INFORMATION TO PREDICTION OF LAW SCHOOL PERFORMANCE
Author(s) -
Reilly Richard R.
Publication year - 1971
Publication title -
ets research bulletin series
Language(s) - English
Resource type - Journals
eISSN - 2333-8504
pISSN - 0424-6144
DOI - 10.1002/j.2333-8504.1971.tb00811.x
Subject(s) - moderation , stepwise regression , selection (genetic algorithm) , test (biology) , psychology , class (philosophy) , index (typography) , regression analysis , mathematics education , rank (graph theory) , statistics , econometrics , computer science , mathematics , social psychology , artificial intelligence , biology , paleontology , combinatorics , world wide web
Graduate and professional schools usually base their admissions decisions on a combination of test scores and on an overall index of undergraduate achievement such as cumulative grade‐point average or rank‐in‐class. The present study sought to investigate whether considering more specific indices of undergraduate performance, through a detailed analysis of college transcripts, could lead to increased accuracy in predicting law school performance. Stepwise multiple regression analyses, performed separately for each of two schools, resulted in the selection of the LSAT, social science GPA, and a moderator variable as the most useful predictors. Because of the effects of selection, however, caution was urged in interpreting the results.